Re: [PATCH 2/2] btrfs: test decompression in the middle of large extents

[Eryu Guan <eguan@redhat.com>]

On Thu, Feb 16, 2017 at 06:32:49PM -0800, Omar Sandoval wrote:
> From: Omar Sandoval <osandov@fb.com>
> 
> This is a regression test for "Btrfs: fix btrfs_decompress_buf2page()".
> It fails for zlib on v4.10-rc[1-7].
> 
> Signed-off-by: Omar Sandoval <osandov@fb.com>
> ---
> This runs in <60 seconds on my test VM, which I think qualifies for the
> quick group?

quick test usually takes <30s, so probably it's not quick enough :)
(it takes ~80s on my test vm, which has 4vcpus and 8G memory)

> 
>  common/btrfs        |  8 +++++
>  tests/btrfs/137     | 96 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>  tests/btrfs/137.out |  2 ++
>  tests/btrfs/group   |  1 +
>  4 files changed, 107 insertions(+)
>  create mode 100755 tests/btrfs/137
>  create mode 100644 tests/btrfs/137.out
> 
> diff --git a/common/btrfs b/common/btrfs
> index 96c3635b..015ce4d2 100644
> --- a/common/btrfs
> +++ b/common/btrfs
> @@ -364,3 +364,11 @@ _reload_btrfs_ko()
>  	modprobe -r btrfs || _fail "btrfs unload failed"
>  	modprobe btrfs || _fail "btrfs load failed"
>  }
> +
> +_btrfs_compression_algos()
> +{
> +	echo zlib
> +	if [ -e /sys/fs/btrfs/features/compress_lzo ]; then
> +		echo lzo

If new compression algorithms are added, are there going to be new
features/compress_$algo entry added? If so I'd suggest that take a more
generic way to query supported compression algos, rather than
"hard-coded" zlib and lzo.

> +	fi
> +}
> diff --git a/tests/btrfs/137 b/tests/btrfs/137
> new file mode 100755
> index 00000000..c3f28cc0
> --- /dev/null
> +++ b/tests/btrfs/137
> @@ -0,0 +1,96 @@
> +#! /bin/bash
> +# FS QA Test 137
> +#
> +# Test decompression in the middle of large extents. Regression test for Linux
> +# kernel commit 6e78b3f7a193 ("Btrfs: fix btrfs_decompress_buf2page()").
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2017 Facebook.  All Rights Reserved.
> +#
> +# This program is free software; you can redistribute it and/or
> +# modify it under the terms of the GNU General Public License as
> +# published by the Free Software Foundation.
> +#
> +# This program is distributed in the hope that it would be useful,
> +# but WITHOUT ANY WARRANTY; without even the implied warranty of
> +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> +# GNU General Public License for more details.
> +#
> +# You should have received a copy of the GNU General Public License
> +# along with this program; if not, write the Free Software Foundation,
> +# Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
> +#-----------------------------------------------------------------------
> +#
> +
> +seq=`basename $0`
> +seqres=$RESULT_DIR/$seq
> +echo "QA output created by $seq"
> +
> +here=`pwd`
> +tmp=/tmp/$$
> +status=1	# failure is the default!
> +trap "_cleanup; exit \$status" 0 1 2 3 15
> +
> +_cleanup()
> +{
> +	cd /
> +	rm -f $tmp.*
> +}
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +
> +# remove previous $seqres.full before test
> +rm -f $seqres.full
> +
> +# real QA test starts here
> +
> +_supported_fs btrfs
> +_supported_os Linux
> +_require_scratch

Better to have

_require_btrfs_command "property"

Not a big deal, but nice to have :)

Thanks,
Eryu

> +
> +algos=($(_btrfs_compression_algos))
> +
> +_scratch_mkfs >>$seqres.full 2>&1
> +_scratch_mount
> +# Need 1GB for the uncompressed file plus <1GB for each compressed file.
> +_require_fs_space $SCRATCH_MNT $((1024 * 1024 * (1 + ${#algos[@]})))
> +
> +# Checksum a piece in the middle of the file. This hits the unaligned case that
> +# caused the original bug.
> +do_csum()
> +{
> +	dd if="$1" bs=4K skip=555 count=100 2>>$seqres.full| md5sum | cut -d ' ' -f 1
> +}
> +
> +# Create a large, uncompressed (but compressible) file.
> +touch "${SCRATCH_MNT}/uncompressed"
> +$BTRFS_UTIL_PROG property set "${SCRATCH_MNT}/uncompressed" compression ""
> +_ddt of="${SCRATCH_MNT}/uncompressed" bs=1M count=1K 2>&1 | _filter_dd
> +
> +csum="$(do_csum "${SCRATCH_MNT}/uncompressed")"
> +
> +for algo in ${algos[@]}; do
> +	echo "Testing ${algo}" >> $seqres.full
> +
> +	# Copy the same data, but with compression enabled.
> +	touch "${SCRATCH_MNT}/${algo}"
> +	$BTRFS_UTIL_PROG property set "${SCRATCH_MNT}/${algo}" compression "${algo}"
> +	dd if="${SCRATCH_MNT}/uncompressed" of="${SCRATCH_MNT}/${algo}" bs=1M 2>&1 | _filter_dd
> +
> +	# The correct data is likely still cached. Cycle the mount to drop the
> +	# cache and start fresh.
> +	_scratch_cycle_mount
> +
> +	# Check the checksum.
> +	compressed_csum="$(do_csum "${SCRATCH_MNT}/${algo}")"
> +	if [ "${compressed_csum}" != "${csum}" ]; then
> +		echo "Checksum mismatch for ${algo} (expected ${csum}, got ${compressed_csum})"
> +	fi
> +done
> +
> +echo "Silence is golden"
> +
> +status=0
> +exit
> diff --git a/tests/btrfs/137.out b/tests/btrfs/137.out
> new file mode 100644
> index 00000000..eb48dd21
> --- /dev/null
> +++ b/tests/btrfs/137.out
> @@ -0,0 +1,2 @@
> +QA output created by 137
> +Silence is golden
> diff --git a/tests/btrfs/group b/tests/btrfs/group
> index ea88ba4e..0a119e4b 100644
> --- a/tests/btrfs/group
> +++ b/tests/btrfs/group
> @@ -139,3 +139,4 @@
>  134 auto quick send
>  135 auto quick send
>  136 auto convert
> +137 auto quick compress
> -- 
> 2.11.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe fstests" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html